FAQ

Home > FAQs > OpenCues vs Compose AI

OpenCues vs Compose AI: what's the difference?

Compose AI proved the inline-trigger idea in the browser: ghost-text completion as you type plus // commands for generation, in Chrome text fields. OpenCues is the same bet rebuilt open source, on any model, and beyond the browser.

What Compose AI does well

Sentence completion with Tab-accept and typed generation commands, years before most tools had either. Its // is a direct ancestor of typed in-field triggers.

Where Compose AI struggles

It lives only in browser text fields, on one fixed cloud model with no key or local option, and the extension's last update shipped in January 2025, so its editor compatibility ages as the web's rich-text editors move. Accepted completions are plain text: no revert, no spans, no transform of text outside the browser.

How OpenCues differs

Scope and model freedom. Compose AI is Chrome-only on a fixed cloud model with a subscription; OpenCues spans Chrome, terminals, coding agents and the shell, is Apache-2.0 and free, and runs on ten providers, your existing AI plan, or local Ollama. Its trigger also does more than generate: transforms, live data and settings ride the same _.

OpenCues edits also land revertable per word, and identity values are tokenised before any cloud call.

OpenCues is also extensible by design: cues and blanks are markdown files you write yourself, and blanks can bind shell scripts or runtime classes, all under an open standard anyone can implement. See write your own cue and write your own blank.

Related: Full comparison · OpenCues on Gmail · Compose AI